|
Activity Number:
|
361
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods*
|
| Abstract - #301631 |
|
Title:
|
Developing A Model For Nonresponse Bias Analysis
|
|
Author(s):
|
Maxime Bokossa*+ and Sameena Salvucci and Dhirendra Ghosh and Marilyn McMillen and Steven Kaufman and Daniel Kasprzyk
|
|
Affiliation(s):
|
Synectics for Management Decisions, Inc. and Synectics for Management Decisions, Inc. and Synectics for Management Decisions, Inc. and National Center for Education Statistics and National Center for Education Statistics and National Center for Education Statistics
|
|
Address:
|
1901 N. Moore St., Suite 900, Arlington, Virginia, 22209, USA
|
|
Keywords:
|
Minimum Response Rate ; Frame Characteristics ; Total Response Rate
|
|
Abstract:
|
A non-response bias analysis was conducted for each of the components of the National Center for Education Statistics (NCES) 1999-2000 Schools and Staffing Survey (SASS). The paper consists of two phases. Phase 1: the results of a descriptive analysis of response rates for all the components of SASS, and statistical tests of response rates for non-uniformity across levels of the frame characteristics as a general indicator of the potential for nonresponse bias. Phase 2 of the paper describes additional analyses on the SASS components that failed to meet NCES minimum response rate requirements in the first phase. Here the paper looks at the affect of the bias on the expected confidence of the confidence intervals and presents a comparison of final SASS estimates with other sources. Also presented is a bias analysis at the item level using variables common in SASS and CCD or PSS. For items with the largest nonresponse only, the paper compares the characteristics of respondents to the item with the characteristics of the nonrespondents to the item among the respondents to the survey.
|
